The growing number of medications used to treat attention-deficit/hyperactivity disorder (ADHD) raises important questions about whether different medications have similar or different therapeutic mechanisms of action. We have recently shown that the stimulant methylphenidate (MPH) and the non-stimulant atomoxetine (ATX) produce clinical improvement via a common mechanism in motor cortex, and distinct actions in frontostriatal and midline cingulate-precuneus regions. These exciting findings offer a window into the common and unique neurophysiological mechanisms of response to stimulant and non-stimulant treatments. However, the interpretation and clinical utility of these results would be greatly enhanced by in- depth investigation of the impact of the two treatments on relevant neural networks, and analyses which evaluate whether improvement is achieved via normalization or other adaptive changes in brain function. The specific aims of this project are to use functional magnetic resonance imaging (fMRI) to determine the significance of activation changes over treatment related to clinical improvement, and the impact of treatment on neural connectivity within and between the anti-correlated frontostriatal 'task-positive' circuit and cingulate-precuneus 'task-negative' network. Our central hypotheses are that clinical improvement is associated with: (i) normalization of reduced connectivity of regions within the 'task-positive' network, with resultant increased inhibition of motor cortex, and (ii) normalizatio of low task-related connectivity in regions within the task-negative network for MPH and the 'task-positive' network for ATX. This research will test a model which posits a neurophysiological basis of mechanisms of response to stimulant and non-stimulant medications, in keeping with our long term objective to develop individualized strategies for ADHD treatment. Testing this model requires large samples of youth scanned using fMRI before and after treatment, and matched healthy controls also scanned twice. We will use an innovative network-based approach to study the effects of treatment, building on results from our current fMRI treatment study, and incorporating new theoretical approaches to understanding ADHD and its treatment. These aims are directly relevant to the NIMH mission as they involve translating basic science findings regarding the impact of different ADHD treatments across multiple brain regions and networks, and exploring whether treatment produces beneficial effects through normalization of underlying pathophysiology and/or compensatory changes. The proposal is innovative as it tests a new model regarding differential effects of stimulant and non-stimulant treatments, and its analytic approach integrates clinical improvement and changes in brain activation. The proposal is important, as validating this model potentially represents an important first step in developing an individualized approach to treatment selection in youth with ADHD, in keeping with the NIH agenda highlighting personalized approaches to clinical care.